US2458601A - Method of and apparatus for measuring pressure - Google Patents

Description

Jan. 11, 1949. J. B. JOHNSON 2,458,601

METHOD OF AND APPARATUS FOR MEASURING PRESSURE Filed March 24, 1944 7 Sheets-Sheet 1 Jan. 11, 1949. J. B. JOHNSON 2,458,501

METHOD OF AND APPARATUS FOR MEASURING PRESSURE Filed March 24, 1944 "7 Sheets-Sheet 2 FIGS 4 imwlal N 5 N TOR J B. JOHNSON ATTORNEY Jan. 11, 1949. J. B. JOHNSON METHOD OF AND APPARATUS FOR MEASURING PRESSURE 7 Sheets-Sheet 5 Filed March 24, 1944 FIG. 4

OOOQOOQOOQQOOOO III E a /5 III E E OOOOQOOOOOOODOQ //v l/EN TOR J B. JOHNSON w. M M T T A V i V B J. B. JOHNSON METHOD OF AND APPARATUS FOR MEASURING PRESSURE Filed March 24, 1944 Jan. 11, 1949.

7 Sheeis-Sheet 4 //V l/E N 7' OR J B. JOHNSON ATTORNEY J n- 1 1949- J. B. JOHNSON 2,453,601

METHOD v0F AND APPARATUS FOR MEASURING PRESSURE Filed March 24, 1944 7 Sheets-Sheet 5 SPACE OF THE ORDER OF A MIL BETWEEN ADJACENT FACE! OF HOLDER 4 SOCKET PLATE TO RM/T FLOW OF AIR.

III/II III/I1 'III/ lNVENTOR J B. JOHNSON A TTORNEV Jan. 11, 1949. JQHNsQN 2,458,601

METHOD OF AND APPARATUS FOR M suRINq pnmssunm Filed March 24, 1944 7 Sheets-Sheet 6 O/L I TRAP TO PUMP SIGNAL 8 ATMOSPHERIC PRESURE '2, I S 2 I METAL (us/m: 1! 9, F l6. l5 :3 22

2 E 0 I8 12 I6 3 l4 HALFATMOSFHERIC PRESSURE 8 :2 Q H] 5 52 677? Y Y E j I INVENTOR E z E JB. JOHNSON I l OZ4GBIDIZIAIGIBZOIZZOZGZ63O BY r/m: (MINUTES) A TTORWE V Jan. 11, 1949. J. B. JOHNSON METHOD OF AND APPARATUS FOR flmsumue rnmssunm Filed March 24, 1944 '7 Sheets-Sheet 7 F IG. /3

F IG. /2

MIN/MUM SPARK/N6 VOLTAGES (KILOVOLTS) M/M SPARK VOLTAGE Le. FOR ZERO ,ua

MIN. VOLTKGE TO GIVE 1014a w M R w a 6 4 2 0 IONIC CURRENT (MICROAMPS) N v, W 7% 0 N0 T %J r M A W J /Z y Patented Jan. 11, 1949 METHOD OF AND APPARATUS FOR MEASURING PRESSURE John B. Johnson, Maplewood, N. J.,.assignor to Bell Telephone Laboratories,

Incorporated,

New York, N. Y., a corporation of New York Application March 24, 1944, Serial No. 528,011

7 Claims.

This invention relates to a method of and apparatus for measuring pressure and more particularly to a method of and apparatus for measuring pressure within hermetically sealed devices.

An object of the invention is to facilitate the measurement of pressures within closed containers.

A more specific object of the invention is to facllitate the measurement of pressures which are lower than atmospheric pressure but not lower than a pressure of approximately one-half inch of mercury.

Another specific object of the invention is to facilitate testing hermetically sealed devices for leakage.

Still another specific object of the invention is to prevent undesired sparking between two oppo sed electrical conductors which are positioned in an area of relatively low pressure.

Quartz piezoelectric plates, when designed for use in apparatus which is likely to be subjected to rapidly changing atmospheric conditions such, for example, as radio apparatus for aircraft, are commonly hermetically sealed, at atmospheric pressure, within a suitable container, which may comprise a box and cover of suitable insulatin material, such as Bakelite, together with a gasket and screws for clamping the cover in closed and sealed position after the plate has been mounted in the holder. Suitable terminal Pins may be mounted on the outside of the box to permit making electrical connection to the electrodes of the crystal plates mounted within the holder. (In order to facilitate the use of piezoelectric crystal plates in electrical circuits, it is usual to provide electrodes on one or more of the surfaces of the plate to which the external circuit conductors may be electrically connected in a suitable manner. These electrodes, for example, may comprise a relatively thin coating of suitable metal such as gold or silver applied, for example,

by the process of thermal evaporation directly to the surfaces of the plate. Again, they may take the form of suitably shaped, separate metal plates held in contact with portions of the surface of the plate by suitable means, such as springs.) One purpose of so mounting the plates is to prevent the possible condensation of moisture on the surfaces thereof due to rapidly fluctuating external atmospheric conditions (pressure, temperature, humidity) such as are likely to be encountered during the flight of the plane. The operation of the plate would be seriously interfered with if droplets of moisture were to form an the surfaces of the plate due to such condensa- During manufacture of the hermetically sealed crystal units, it is necessary of course to test the completed, sealed unit in order to determine the effectiveness of the seal; the method and apparatus of the present invention are particularly applicable in connection with the testing of such units and the invention will be described as so applied. It will be understood of course that the invention is not limited in its application to this particular use.

A feature of the present invention is means, surrounding the exposed portions of the terminal pins of the sealed holders being tested, for preventing sparking between such exposed portions.

There is disclosed in the copending application Serial No. 528,013, filed March 24, 1944 in the name of G. W. Willard, entitled Method'of and apparatus for measuring pressure a novel arrangement for testing sealed containers which is based on the discovery that the gas pressure around two opposed or separated electrodes, 1. e., electrical conductive members or elements, may be determined by measuring current flow, at constant voltage above-the minimum sparking voltage, between two electrodes, 1. e., the method depends upon the variation of an ionization current between a pair of electrodes (for a given voltage) with changes in the pressure of the gas surrounding the electrodes. So far as the range of pressures in which the arrangement is designed to function is concerned, the ionic current increases as the pressure of the gas decreases.

The arrangement of the present invention is applicable particularly in connection with the testing method and apparatus disclosed in the copending application referred to and will be so described herein.

In accordance with a specific embodiment of the arrangement referred to above, test equipment designed to test a plurality'of hermetically sealed quartz plate holders comprises a cabinet on which is mounted an evacuation chamber. A vacuum pump for evacuating the chamber to the desired level, a pressure regulator, pressure gauges, voltage supply means, current meter and other incidental items of equipment are provided certain of which may be mounted on or within the cabinet. A socket plate is mounted in the lower portion of the evacuation chamber for removably supporting the sealed crystal holders which are to be tested. In use the holders to be tested are mounted on the socket plate within the evacuation chamber, 'the cover thereof is aerator 3 closed and sealed there y completing a gas-tight enclosure around the holders and the pressure within the chamber is lowered to the particular level predetermined in accordance with the requirements of the test. After maintaining this level for a predetermined time. a regulated, i. e...

a current-limited, voltage is applied across the terminals of each sealed holder in turn and the corresponding ionic current flow (if any) between the separated or opposed electrodes within the respective holder is measured. The intensity of the ionic current flow will be an indication of the pressure of the air or other fluid within the respective holder and it may be readily observed therefore how much, if any. the pressure within the holder has dropped due to leakage through the seal to the surrounding area of lowered pressure.

Magnitudes of pressure will be defined herein by the terminology commonly followed in the scientific and technical publications, 1. e., the term a pressure of eight inches of mercury," for example, means a pressure that will Just support a column of mercury eight inches in height; "a

pressure of one-half inch of mercury means a pressure thatwill Just support a column of onehalf inch in height and so on. Similarly the term ambient pressure is used in its commonly accepted sense to define the pressure immediately surrounding the object or area in question. The term gas is used in the usual scientific sense and is intended to include atmospheric air. While the term hermetically sealed" is used at various points herein in reference to containers the seals of which are being tested to determine whether or not the seal is effective to meet specified requirements which in some instances may permit a slight leakage, it is of course understood that, in the exact sense, the term "hermetically sealed" should be applied only when the seal is perfect, that is, when it will entirely eliminate leakage between the interor of the container and the surrounding area. For purposes of the present description, the term ionic currentf should be taken as inclusive of current arising through "breakdown between separated portions of the physical conductors resulting, for example, in sparking or current across space but noninclusive of current flowing only in the physical conductors themselves or between the elements of a battery or other potential source not involvng gaps or spaces.

Fig. 1 is a front elevational view of an arrangementfor testing hermetically sealed containers which embodies features contemplated by the present invention;

Fig. 2 is a side elevatlonal view of the device illustrated in Fig. 1;

Fig. 3 is a plan view of the device illustrated in Fig. 1;

Fig. 4 is a perspective view of the cabinet of the arrangement of Fig. 1 with the front panel and associated apparatus mounting base removed and with the top panel in raised position to show the underneath side thereof;

Fig. 5 is a perspective view of the front panel and associated'mounting base taken from the rear;

Fig. 6 is a view in perspective of the evacuation chamber with the cover in raised position to disclose a number of crystal plate holders mounted in position for testing;

Fig. 7 is an enlarged sectional view of a portion of the socket plate taken on line 1-1 of- Fig. 6, showing particularly the arrangement of 4 thesprlngiackterminalsinthesocketplate and illustrating the novel means for preventing sparking contemplated by the invention;

Fig. 8 is a view in perspective of one of the spring jack terminals;

Pig. 9 is a sectional view of a sealed crystal plate holder of a type that may be tested by the contemplated arrangement, a piezoelectric crystal plate being shown mounted within the holder;

Fig. 10 is an end sectional view of the crystal plate holder of Fig. 9;

Fig. 11 is a schematic showing of the electrical circuit utilized by the testing arrangement;

Fig. 12 is a curve illustrating the relationship between pressure and ionic current for a number of different voltage values;

Fig. 13 is a view of the mercury manometer switch;

Fig. 14 is a curve illustrating leakage rates for three difierent magnitudes of leak; and

Fig. 15 is a sectional view of a form of auxiliary spark gap.

Referring now to the drawings, there is illustrated a testing device comprising a cabinet i I which is provided with a removable front panel l2 and removable top panel or cover plate [3. As shown particularly in Fig. 5, front panel I2 and the associated apparatus mounting base l4 may be removed as a unit from the cabinet. Plugtype connectors l5 and I6 within the cabinet H (see Fig. 4) engage the jack-type connectors l1 and 2|, respectively, (Fig. 5) when base I4 is in position within cabinet ll thereby maintaining electrical continuity between equipment carried by the removable unit and equipment carried by the cabinet proper. Similarly, top panel i3, as

indicated most clearly by Fig. 4, may be readily V removed as a, unit from the cabinet; plug assembly l8 and jack assembly l9 cooperate when panel I3 is in closed position on the cabinet to maintain electrical continuity between the respective equipment elements.

As shown particularly in Fig, 5, several equipment units are mounted on base l4 and front panel l2. These units. the function of which will become apparent from subsequent consideration of the schematic circuit and other portions of the specification, include condenser 23, vacuum tube 24, transformers 25, 26 and 21, relay 4i,

fuses 42 and 43 with suitable fuse holders, voltage adjusting autotransformer 44 and resistance element selecting switch 45. The resistance elements are preferably mounted directly on the two plates of switch 45, several of the elements being shown in the drawing. As shown in Fig. 1, dials 46 and 41 are provided for operation of voltage and resistance selectors 44 and 45, respectively. Oil-on switches 5i and 52 are also mounted on panel l2.

Evacuation or vacuum chamber 53 is mounted on top panel l3, being shown with cover 54 in closed position in Figs. 1, 2 and 3 (in which position a gas-tight enclosure is formed) and with the cover in open position in Fig. 6. Cover 54 is supported, somewhat loosely, by hinge arms 55 and 55 which are rotatably attached by suitable bolts or other pivots to upright members 51 and II, respectively. Members 51 and H, together with top cross bar I2, act as a support for cover 54 when it is in raised or open position. As apparent from Figs. 2 and 3, members 51 and II extend down the rear of cabinet I I to the base thereof and act also as a guard for exhaust tube 13.

As shown clearly in Figs. 6 and 7, evacuation chamber 53 comprises, in addition to cover 54, a socket plate I4 of suitable insulating material and base-plate 15 also of suitable insulating material, gasket 10 being positioned between the two plates. A second gasket I1 is positioned on the exposed upper surface of plate 15 and serves to seal the chamber when cover 54 is in closed position.

As shown in Fig. 6, the socket plate is provided with a plurality-of pairs of apertures, sockets, or holes, the number of pairs of holes determining the number of sealed holders or containers that can be tested at one time. (The capacity of the set illustrated, therefore, is forty-four holders although the capacity may be made greater or less as desired.) The two holes of each pair are properly spaced to receive the two terminal pins of a respective one of the holders being tested. A number of hermetically sealed holders, as 8| and 82, are shown in position for testing in Fig. 6.

A spring clip or jack terminal assembly of the type shown in Fig. 8 is associated with each of the holes of the socket plate. Each terminal assembly comprises a socket or lack portion 83, a shoulder 84 and a threaded bolt portion or stud 85. As shown in Fig. 7 a soldering lug is positioned on each bolt portion of the terminal assemblies, being held in place by a nut. For example, soldering lug IOI is held on bolt 85 by nut I02. These soldering lugs facilitate the electrical connection of conductors (leading to apparatus to be described subsequently) to the respective bolts or studs and thence to the socket portions of the terminal assemblies. Conductors I03, I04 and I05 are shown in part in Fig. '7.

It will be apparent from Figs. 7 and 4 that the spring contact or terminal assemblies act not only as electrical connecting members but also as mechanical elements for sealing the holes in base-plate I5, shoulders, such as shoulder 84, being eifective for this purpose. Socket plate I4 is held in its proper registering position on baseplate 15 by a number of suitable screws. screw 86 being shown in Fig. '7. Socket plate I4 may be easily removed by removal of these screws without disturbing the seal effected between gasket I6 and base-plate I5 by the shoulders (such as shoulder 04). The novel construction is such that the evacuation chamber proper is sealed by gasket I6 so that when cover 54 is in closed position and is sealed by gasket 11 and when the chamber is evacuated through exhaust tube I3, the "base of the chamber so far as changes in atmospheric pressure are concerned comprises gaskets I0 and 11. The shouldered portions provided on the terminal assemblies, as shoulder 84, are effective in bringing this sealing action about.

The novel construction of the evacuation chamber is such that the socket plate may be easily removed for cleaning and the stud, or bolt, mounting may be easily replaced. Further the novel form of contact terminals utilized and the spacing thereof in the socket plate permit testing of holders having slightly different types of pin terminals. In the event that it be desired to test containers having a terminal, arrangement substantially different than that of the holders illustrated, top panel I3, which carries socket plate I4 and associated wiring as well as the switches and meters referred to, above, may be easily removed and replaced by another top panel provided with a socket plate arranged to accommodate the particular terminals involved. The provision of the jack and plug-type connection (plugs I8 and jacks I9, Fig. 4) referred to above facilitates the removal of one top panel and replacement thereof by another; (meter- I01 and associated switch I'll are preferably lack mounted so that they may be readily removed from one panel and replaced on another). 1

It will be evident that with holder 0| in position as indicated in Fig, 7, conductors I03 and I04 will be electrically connected through jack terminals 03 and I08, respectively, to terminal pins I05 and I01, respectively, of the holder; terminal pins I05 and I0I in turn are electrically connected to the respective electrodes for the crystal plate mounted within the holder.

As mentioned above, evacuation chamber 53, when closed and sealed by cover 54, may be exhausted to the desired pressure level through exhaust tube 13. The apparatus utilized in bringing the pressure within the evacuation chamber to various desired levels (by a method which will be described in detail subsequently) includes a vacuum pump III driven by motor II2. Pump III is connected through tube I I3 to oil trap I I4, entering the latter through the base thereof. Exhaust tube I3, which, as stated above, is connected to the evacuation chamber 53, is also connected to oil trap I I4 entering the latter through the top cover thereof.

Gage H5, which may be one of the ordinary types of steam vacuum gage and which is connected to oil trap II4 through a suitable tube or pipe leading through the top cover thereof, provides an immediate indication of evacuation when pump III is started and also gives a rough indication of the pressure within the system so long as the pressure is not too low (where barometric changes have an excessive effect on the readin of such a gage) say above a pressure of approximately ten inches of mercury. Such a gage of course reads pressurees below atmospheric, rather than the absolute pressure. For accurately measuring and indicating pressures below 10 inches of mercury, mercury manometer H6 is provided. Manometer H6 is connected to oil trap 4 by tube II! which enters through the top cover thereof. Back plate I3I, which is attached to the rear of cabinet II, acts as a support for manometer I I0 and its scale.

A pressure control auxiliary is provided for the purpose of keeping the absolute pressure in the evacuation chamber from going below a predetermined value, for example, below a pressure in the neighborhood of one-half inch of mercury. (The purpose of such a control will be brought out subsequently.) This control auxiliary includes closed mercury manometer switch I32 (Fig.

' 13) which is controlled by absolute pressure and which operates through holding relay 4|, (Fig. 5) solenoid-operated valve I34 which in turn controls the admission of air through inlet pipe I33. Inlet pipe I33 enters oil trap I I4 through the bottom thereof. An adjustable needle valve I35 is associated with solenoid valve I34.

As indicated most clearly in Figs. 13 and 11 manometer switch I32 may be adjusted to approximately maintain any desired pressure within a certain range by adjusting its tilt.

When the pressure within evacuation chamber 53 falls below the'predetermined minimum value, for example one-half inch of mercury, valve I34 opens and allows air to leak into the system through inlet tube I33 at a rate which is controlled by adjustable valve I35 and which is slightly faster than the pumping rate, until the pressure rises to apredetermined maximum value,

for example, five-eighths of an inch of mercury,

repeated Releasevalvelflismountedonthersaroi evacuationchamberlhthisvalvemaybeopened upon completion of a test to permit air to enter evacuation chamber l2 in order to bring the interior thereof back to atmospheric pressure, be-' fore opening the cover.

Four double-tier selector switches are mounted on top panel I2, the lower switch in each respective pair being shown in Fig. 4. The switches shown are designated respectively I44, I44, I42 and I41. A knob is provided for rotating each pair of switches, these knobs designated III, I22, I42 and I24 being shown in Figs. 1 and 3. As shown most clearly in Fig. 11 and as indicated in Fig. 4, each of the switches oi one set has one terminal connected to one common conductor of the testing circuit while each of the switches of the other set has one terminal connected to a second common conductor of the circuit. Thus switches I44, I45, I48 and I41 of the "lower set each have a terminal connected to conductor I" while the switches oi the "upper" set each have a terminal connected to conductor I62. Corresponding individual switching terminals of the switches are connected to the respective pairs of spring, clip or jack terminals as indicated in Fig. 4.

In the embodiment of the invention selected ior disclosure wherein the socket plate is designed to mount forty-four holders arranged in four rows of eleven holders each, each set of selector switches is arranged for control oi one of the four rows and each switch is provided with twelve terminals, one terminal in each case corresponding to the "off" position of the respective switch, the other eleven corresponding to the eleven holders to be tested.

Microammeter I61, together with switch "I for shorting the circuit beyond the meter (1. e.,

assassishorting leads I55 and I66, Fig. .11) is mounted on top oi panel I3. Lamp indicators I12 and I13 are mounted on front panel I2 and associated respectively with switches BI and 52. Hand grips I14 and I15 are also mounted on the front panel to facilitate removal and replacement of the same. Fuses I16 and I11 may be replaced from the front of the panel (Fig. 1). 1

For purpose of further describing the arrangement and operation of the invention, it will be assumed that a number of hermetically sealed crystal plate holders of the general type illustrated in Figs. 9 and 10 are to be tested. The holder selected for illustration is of the so-cailed "pressure type, i. e., piezoelectric plate III is positioned between electrodes I82 and I83, which are in the form of metal plates, and firm contact between the piezoelectric plate and the electrodes is assured by spring I84 while a second spring I" aids in maintaining the assembly in proper position within, the holder. Terminal'. pins I06 and I01 are connected to the respective electrode plates by suitable means. Shoulder portions I25 and I81 are provided on respective'pins III. and I01. The holders are placed in position on socket plate 14, as shown in Fig. 6, the respective terminal -pins oi the holders being positioned in the proper apertures in the socket plate. (Fortyfour holders may be mounted in the socket plate illustrated but it will be understood that socket plates with greater or less capacity may be utilized. as desired.)

After the holders to be tested are placed in position, cover 54 is closed against gasket 11, to seal 8 encuationchamberuaeleasevalveiflhckled andevacuatimpmnpiilisstarted. .Theholdsrs shouldbe-carefullyseatedinpositionsothatne" appreciable gap will be allowed between the base oieachholderandthesdiacenttopiaceoithe socket plate. As described subsequently. however. there is a small but distinct passageway for air provided between the holder base and the face of the socket plate so that there will be an occasional air leakage between the upper suriaee of the socket plate and the adjacent suriace oi the holders; in other words, the "base" oi the evacuation chamber proper is defined by gasket 1' and the shoulders of the Jack terminals (as shoulder 84, Fig. '1) and not by the upper suriace oi socketplate 14 and the engaging surfaces of the holders.

Assoonaspump lIiisstartedJflle Illshould indicate (by an increasing vacuum reading) a lowering pressure assuming that a good seal has been attained between cover 54 and gasket 14. If gage II! indicates that the apparatus is operating satisfactorily, the exhausting process continues until mercury manometer IIO indicates an absolute pressure in evacuation chamber 22 of about one-half inch of mercury.

As the pressure in the evacuation system is low ered, the mercury in manometer switch I 22 will fall; 1. e., the level in the right hall of the U tube will drop and the level in the left half oi the tube will rise, Figs. 11 and 13. So long as the level in the right half of the tube is sumcienty high to engage either terminal I or I42, relay M will be held operated by current supplied from the secondary winding of transformer 21 over conductor IQI, operating winding of relay 4!, terminal I22 and back to the other side of the secondary winding of transiormer 21 over either of two paths depending upon the mercury level, i. e., either from terminal I over conductor I92 and from terminal I42 over conductor I94 and make contact of relay H in parallel ii the mercury be high enough to contact terminal I or from terminal I42 over conductor I84 and make contact of relay alone if the mercury be in contact with terminal I42 but out oi contact with terminal I. However, as soon as the level of the mercury in switch I22 drops below terminal I 42, the operating circuit of relay 4I will be broken and the relay will release.

As soon as relay 4I releases, solenoid-operated valve I24 will be operated over a path which includes conductors I95 and I26, break contact 0! relay 4| and conductor I91. Operation oi valve I24 allows entry of air into the evacuation system through inlet pipe I22 thereby raising the pressure within evacuation chamber 52. The rate of entry of the air is governed by the position of adjustable leak valve I25. Asthe pressure in the system is now raised due to admission of air through inlet pipe I22, the mercury will rise in the right-hand half of the U-tube of switch I22 until contact is made with upper terminal I whereupon relay 4| is operated and valve I24 moves back to closed position thereby closing inlet pipe I22. The cycle is then repeated continuously.

It will be noticed that, while the operating circuit of relay H is not completed until the mercury reaches upper terminal "I, i. e., it is not completed when the mercury reaches the lower terminal I42, the relay, once operated, will be held operated until the mercury drops below the lower terminal I42. The pressure is thereby automatically fluctuated in the range between a selected maximum pressure, which may be, for example,

9 five-eighths of an inch of mercury, and a selected minimum which may be, for example, one-half inch of mercury.

desired to obtain a pressure lower than the pre-- determined minimum," needle valve I35 may, of course, be closed temporarily.

After chamber 53 has been evacuated to the desired low pressure and has been maintained at such low pressure a period of time determined by the particular testing instruction, the testing of the individual holders is started. (The period of testing time is normally measured from the moment at which solenoid switch first operates, i. e., about one minute after starting the pump.) The test voltage is turned on and regulated to the desired value by operation of voltage regulating transformer 44 adjusted by dial knob 46. The value of the applicable voltage may be read by closing switch III momentarily and noting the reading on microammeter I61; it will be assumed that transformer 44 is adjusted for a reading of 30 microamperes on meter I01 when switch I'll is closed. For such adjustment of transformer 44 any one of five definite voltages maybe obtained depending upon the position of selector switch 45 operated by associated dial knob 41; switch 45 has five positions, labeled 2 to 6 on the dial, and is shown in Figs. 1 and 11 in position No. 3. It will be assumed that with the adjustment of transformer 44 referred to, voltages of 600, 900, 1200, 1500 and 1800 volts, respectively, are obtained with the five settings of switch 45, since the respective switch settings introduce into the circuit respective resistances of 20, 30, 40, 50 and 60 megohms. These resistance elements are effective to limit the current passed through the holders to a maximum value of 30 microamperes.

The first of the forty-four holders to be tested is now selected by operation of the proper pair of the four pairs of selector switches I44, I45, I46 or I41 controlled by knobs IBI, I62, I63 and I64, respectively. (These selector switches are left in the normal off position except when actually in use for selecting a holder in the row controlled by a respective switch.) It will be assumed that the first holder to be tested is holder 8| which occupies the first position in the row of eleven holders controlled by lower selector switch I44 and upper selector switch 203 (Fig. 11) of the same set. Switches I44 and 203 are moved to position I therefore by rotation of dial IBI thereby connecting testing circuit conductor I66 to conductor I03 and testing circuit conductor I65 to conductor I04, respectively. It will be apparent from reference to Figs. 4, '7, 9 and 11 that the testing voltage is now applied across terminal pins I06 and I! of holder 8| which is being tested. Terminal pin I01 of the crystal holderis connected by a suitable conductor provided within the holder to one electrode I82 of crystal plate I8I (Fig. while terminal pin I06 is connected to the other electrode plate I83.

Now, as set forth more in detail in the copending application referred to above, it has been discovered that in a case such as this there is a definite relationship, changing as the applied regulated voltage is changed, between the ionic current flow (if any) across the gap between the two electrodes of the crystal plate and the gas pressure within the holder and surrounding the plate. This relationship, voltages and for a specific type of holder and electrode arrangement, is shown by the curves of Fig. 12. For example, assuming a test voltage of 1200 volts, it will be noticed that the ionic current is zero for pressures of twelve inches of mercury and over, and gradually increases as the pressure drops below twelve inches, reaching a maximum of approximately 18.5 microamperes at a pressure of approximately one-half inch of mercury.

It will be noticed from the curves of-Fig. 12 that once a critical" pressure in the neighborhood of one-half inch of mercury has been reached, any further decrease in pressure results in an abrupt drop of the ionic current to zero. This is the reason for the provision of the arrangement described above whereby the pressure within the evacuated chamber is held above a predetermined minimum value. In other words, the arrangement is designed to operate in a pressure range whichis not permitted to fall below approximately one-half inch of mercury, the exact value depending upon the type of holder and electrode arrangement.

faces of the metal'electrode plates I82 and I83 (Fig. 10).

It is of course desirable that no ionic current flow be set up between the exposed portions of the terminal pins, 1. e., the portions outside of the holder under test, as this would defeat the attempt to determine the conditions within the holder. It is apparent, however, that these portions must themselves be in the low pressure area since one of the most likely places of leakage is the point at which a terminal pin enters the holder. It is not feasible, therefore, to hermetically seal off these portions of the terminal pins from the low pressure region but it is necessary to provide open passages for the fiow of air from areas near the base of the terminal pins to the region of reduced pressure.

Now, in accordance with the novel features of the present invention, the establishment of ionic current flow between the exposed portions of the terminal pins, which as stated above are in the low pressure region, is prevented in a unique and effective method. Briefly, according to this novel method, the passages provided for flow of air, while of suflicient area for the purpose provided, i. e., to permit the rapid flow of air in the event of leakage of air at the base of the terminal pins, are narrowed or constricted by insulating walls sufiiciently to be effective in preventing the establishment of undesired current flow between the terminal pins;

The underlying theory which according to 219- I plicants present opinion accounts for the successful functioning of the method may be explained something as follows. Electric breakdown in a narrow passage between insulatin walls is impeded and the likelihood of its occurring is lessened by charges that tend to collect for three different test 11 on the walls in such a way as to oppose the fields which are tending to set up the now of current. To explain the theory further, if a pair of ions be formed in the space between two opposed electrodes, 1. e., the two terminal pins in our case, by some cause or other, these ions could normally start a cumulative process of ionization, assuming that the conditions of voltage and pressure are favorable as they are in our present case. Thiscumulative process would ultimately result in the flow of a steady ionization current. However, if now, in accordance with the features of the present invention,'the passage between the electrodes be constricted by insulation the ions will then often strike one of the walls of the passage defined by theinsulation and will adhere thereto instead of passing on to their respective electrodes with cumulative ionization. Ions of one kind adhering to a wall will attract and neutralize ions of the opposite kind and this tends to clear the passage of ions thereby decreasing very materially the likelihood of the current growing up to the steady, self-sustaining condition.

The passage should be very narrow. i. e., of

the order of a mil or so it is diiilcult to establish the exact, most favorable, dimension according to a general rule and it can best be determined by experimentation in each case. Ions movein the passage under the influence of the electric field between the electrodes (terminal pins) but with frequent diverting collisons with molecules of the gas (air). The passage should be so dimensioned that an ion, in its motion, would be much more likely to strike the insulating wall than to traverse the passage longitudinally without striking the-wall.

Considering now the practical embodiment of the above method in the testing apparatus referred to above and referring particularly to Fig. 7, it will be seen that socket plate 14 is provided, in addition to the apertures of proper size to receive the terminal pins, with enlarged portions of each such aperture which are of proper diameter to accommodate the shoulders of the respective terminal pins. Thus, as illustrated, shoulders I" and Ill of terminal pins I08 and I" are seated in the respective enlarged apertures provided when the pins are positioned in the respective sockets 83 and I08. The depth of these enlarged apertures is carefully determined 12 tween'sockets CI and III, for example, would be from one socket up through the associated aperture, 'to the surface of socket plate I4 and down to the other socket through the associated aperture: establishment of ionic current over such a path will be unlikely due botli to the length of the path and to the constric d area encountered at the surface of the soc et plate.

It will be understood. of course, that the socket plate will be designed in accordance with the type of holder that is.t0 be tested; particularly the depth of the enlarged portions of the apertures will be carefully determined in accordance with the area of the shoulders on the terminal pins of the holders. However, in the event that the socket plate is not provided with the enlarged apertures necessary to accommodate the shoulders of a holder to be tested, a spacer of suitable dielectric material. provided with properly spaced and dimensioned apertures to accommodate the shoulders of the terminal pins, may

' be positioned between the socketplate and the holder. Or again, a strip of suitably dimensioned dielectric may be so positioned between the holder and the socket plate as to occupy substantially all the space between the two terminal pins of each respective pair of pins. In the even that during a test, certain positions -on the socket plate are vacant, i. e., not occupied by holders, it may be well to place a sheet of dielectric over the vacant apertures in order to prevent possible ionic current flow from one socket to an adjacent socket in the manner and is such that when the holders are in position on the socket plate a passage for the flow of air will be established between the upper face of the socket plate and the adjacent face with the theory discussed above, to prevent the establishment of ionic current flow between the respective pairs of terminal pins. Actually, as

indicated by the legend of Fig. "l, the space between the face of the socket plate and the face of the holder will be of the order of a few mils only and will not be readily discemable to the observer.

It will be readily apparent from Fig. 7 that the novel arrangement of socket plate I4 is such that not only is the establishment of ionic current between the terminal pins prevented, but the establishment of such current between adjacent sockets, as 83 and I", is also prevented. In view of the manner in which each socket is surrounded by the dielectric, it will be apparent that the only possible path for current flow bereferred to above.

Returning now to consideration of the testing procedure as applied to holder ll it will be apparent from the above that as the testing voltage which we will assume is regulated at 1200 volts, is applied across terminals I" and I01 of the holder, the resulting ionic current flow, if any, across the crystal plate within the holder will be an indication of the pressure within the holder; the intensity of ionic current flow will beindicated by microammeter I", switch Ill being open (Fig. 11).

'-It will be obvious that i: holder II, which it will be remembered was sealed at atmospheric pressure, is imperfectly sealed and subjected to leakage, i. e., if a true hermetic seal was not obtained,- the pressure within the holder will drop, 1. e., air will escape therefrom, during the exhausting of the evacuation chamber and maintenance of the pressure therein at the low point under which the tests are carried out. The amount of leakage from holder II to the lower pressure surrounding it will depend, of course, on how imperfectly the holder has been sealed. As the testing voltage is applied, if no ionic current flow be registered, this will indicate that the pressure within holder 8| is twelve inches of mercury or higher (curve of Fig. 12). However, if ionic current of, for example, 8 microamperes be registered, this will indicate that the pressure within holder 8| has dropped to approximately seven inches of mercury and so on. Interpretation of these results will depend upon the requirements of the particular test, for example, it maybe that all holders-showing any ionic current flow at all will be rejected or it may be that only those showing ionic current flow greater than a predetermined value will be rejected.

After holder 8| has been tested, the selector switch comprising lower" switch I and "upper switch ill will be operated to selectively test the other ten holders in the row. In the same manner the other three sets of selector switches will be operated in turn to test the holders in the other three rows until the entire i'orW-four holders have been tested. A suitable record will be kept during the tests from which the faulty holders may be identified. The entire testing procedure takes very little time as only a few seconds are required for testing each holder, that is, just sufllcient time to observe whether ionic current is passed or, ii. the requirements of the test are such as to necessitate it, the intensity of such current.

Upon completion of the tests, release valve I43 is opened-and pump ii i is stopped. When the vacuum has been'entirely relieved, cover 54 is opened and the holders are removed from socket plate H. Those holders identifiedas defective are then set to one side for repair and resealing.

Before making the above tests, if it be suspected that any of the holders are subject to conduction shorting, i. e., have a low resistance leakage path between their terminal pins through the base of the holder or becauses of contamination on the crystal, such holders should be tested for resistance before evacuation by application of the contemplated test voltage. Any that show current readings at atmospheric pressure cannot be tested for leakage. Conduction shorts that might not be detected otherwise, for example, by use of low resistance ohm meters, will be indicated at the higher voltages used in the present arrangement.

If desired, an audible or visual signal (indicated by box 208, Fig. 11) may be incorporated in the circuit; such a signal which may take the form of a bell or lamp for example, facilitates the detection of defective holders as it makes it unnecessary to watch meter I61. The signal unit may comprise, by way of example, a resistor connected as a series element across which the ionic current will produce a voltage drop, an amplifier for amplifying this voltage drop, a sensitive relay operated by the amplifier output and a visual or audible signal in the local circuit of the "relay. Suitable adjustment of the signal unit may be made in accordance with the testing requirements; i. e., in accordance with the minimum ionic current which it is desired to indicate or detect.

It will be understood, of course, that the particular type of holder illustrated in Figs. 9 and 10 has been selected by way of example only and that the arrangement may be used in testing other types of holders. The absolute pressureionic current curves for different types of holders will be generally similar to that of the pressure type holder (Fig. 12) but may differ in values. Data for plotting typical curves for different types of holders may be obtained by making tests of assembled, but unsealed, holders of the respective type involved.

Among the particularly valuable features 0 the testing arrangement described above is the ease with which the conditions under which the holders are tested may be accurately varied in accordance with varying requirements of the testing requirements. For example, the severity of the test may, of course, be increased by lengthening the period of time during which the holders are subjected to the lowered pressure within the evacuated chamber before the ob servational steps of the test are commenced. Also by simple switch ach'ustment any one of a number of different testing voltages may be selected. Further the holders need not be immersed in liquid during the test; many previous testing arrangements utilized immersion in water or other liquid, often, with harmful results.

Another particularly valuable feature of the arrangement resides in the fact that it permits the measurement of magnitude of leaks or leak rates with consequent rejection of holders showing leaks greater than any predetermined allowable leak rate. -In this connection magnitude of a leak is intended to mean the time required for the pressure within a holder to drop from one atmosphere to one-half atmosphere when the holder is immersed in a perfect vacuum, the time being counted from the instant of immersion. (For a stable leak the above might be restated as the time required for the internal pressure to drop to one-half the difference between the external and internal pressures.) For example, if a holder sealed at atmospheric pressure is placed in a vacuum chamber which is evacuated to a negligible pressure in a negligible time, and if the pressure within the holder drops to one-half atmosphere in five minutes the leak is referred to as a five-minute leak. There is illustrated in Fig. 14 curves showing the leakage rates for three different magnitudes of leak, i. e., two and onehalf, five and ten-minute leaks. As clearly shown by these curves, the pressure within a holder with a five-minute leak falls to one-half atmosphere in five minutes, that of a holder with a two and one-half minute leak in two and onehalf minutes and that of a holder with a tenminute leak in ten minutes. Also that the pressure in the first-mentioned holder falls to onequarter atmosphere in ten minutes, to oneeighth atmosphere in fifteen minutes and so on. Curves of this nature may be utilized to advantage in conjunction with the absolute pressureionic current curves of a particular type of holder for setting the time required for leak testing under specific electrical settings.

While the testing arrangement in connection with which applicant's invention is particularly applicable has been described in connection with the testing of hermetically sealed crystal plate holders that have been sealed at atmospheric pressure, it may be used, of course, for testing other types of containers, suitable modifications being made in the socket plate as required. Further, the arrangement may be utilized for testing the seal of containers that have been sealed at pressures other than atmosphericpressure. The arrangement may be used, for example, in testing such sealed devices as electrical condensers and, in short, may be adapted generally to the measurement of degrees of vacuum or pressure within enclosures or in space. The arrangement is intended to function, however, only when the pressure being measured does not fall below one-half inch of mercury. It'is realized that in some instances the separated electrodes providedwithin the container in connection with the normal function of the device may not be adaptable to use in connectionwith the present method and that in other' instances the enclosure may not normally be supplied-with separated electrodes of any nature, and it is contemplated that in such instances a set of electrodes be provided within the container for the express purpose of utilization in the method of the present invention.

In instances such as mentioned in the closing sentence of the above paragraph, i. e., where for one reason or another a set of electrodes is to be provided for the express purpose of pressure measurement, it is obvious that it will be advanta'geous to provide electrodes of a type that will tend to produce a stable and controllable spark gap. Among the types of spark gaps that may be used are those formed by opposed ball electrodes, crossed opposed cylindrical electrodes or opposed rounded plate electrodes; these electrodes may be made of non-sputtering material such as aluminum, or gold may be utilized where long life is particularly desired. Also in the instance of metal enclosures, the auxiliary spark gap may be established between a single sealed-in insulated electrode and the adjacent wall of the container. There is illustrated in Fig. 15, for example, the manner in which a gap may be established between one terminal pin and the adjacent wall of the metal container; the terminal pin may be sealed in the container in the general manner illustrated in Fig. 3b of Patent 2,048,556 issued July 21, 1936 to E. D. McArthur.

While the method of pressure measurement described above involves ionization current values for a particular fixed voltage, it is contemplated that other procedure may be followed. For example, it is contemplated that pressure may also be measured by observing the minimum voltage required to give an ionic-current passage (or spark) reference being had to the minimum sparking voltage" curve of Fig. 12. Further the minimum voltage required to give a current of (n) aa might be observed rather than a current that is just detectable; a curve of Fig. 12 shows minimum voltage to give ac current.

- It will be noticed in referring to the, curvesof Fig. 12 that points A, B and C of the minimum sparking voltage-pressure curve correspond, re-.v

spectively, to points A, B and C, respectively; of the ionic current-pressurecurve and that points D, E'- and F' .of the 10 m curve correspond, respectively, to points D,-E and For the ionic current-pressure curve. I

Although the use of direct current has been described in connection with the testing method,

it willbe understood that alternating current may be used in certain instancesifdesired.

In the event that the holders to be tested have been sealed at other than atmospheric pressure- (it is pointed out above that the testing method is readily applicable in such instances), it must be ascertained and taken into account whether or not the holders remained up to the time of test in the same ambient pressure as that under which they were sealed. If it be established that the holders were subjected to a difi'erent ambient pressure it must be ascertained for how long a period they were so subjected; this procedure is necessary in view of the possibility that air may leak into, or out of,the holder during such period. The following specific procedures may be followed to advantage in testing holders which have been sealed at pressures other than atmospheric pressure.

First, in the event that the holders have been sealed at a pressure above atmospheric and have then been subjected to atmospheric pressure for a time (t), the sealing pressure and the time (t) suming, of course, that time (t) and the dif ference in pressure are sufficient to indicate the possibility of appreciable leakage of air into, or out of, the holder before the test is started.

Secondly, in the event that the holder has been sealed at pressure (Pl) which is below atmospheric pressure but above one-half inch of mercury, the holder as sealed should give a large ionic current. If the holder is now placed at atmospheric pressure, the internal pressure will rise from (P1) in the event that leakage occurs. Such holders may be tested therefore by a method which includes subjecting them to atmospheric pressure for a time (t) and then making the ionic current test at atmospheric pressure, (i. e., not in the vacuum. chamber) during which test the smaller the ionic current the greater will be the indication of leakage. This is a simple and efficient method of testing holders where they can readily be sealed at pressures of from one-half to one inch of mercury.

Third, in the event the holder has been sealed at a pressure (P1) which is below critical pressure (i. e., below one-half inch of mercury where the ionic current is maximum). Now by employing a test voltage sufilciently high to result in ionic current even at atmospheric pressure and by making two or more tests with the holder at atmospheric pressure (as outlined in the paragraph above) it can be determined whether the internal pressure has changed from the range of pressures in which ionic current increases with increasing pressure (pressures of zero to one-half inch of mercury) to the range of pressures in which ionic current decreases with increasing pressure.

It is also contemplated that a holder may be sealed with one gas confined within the holder which is then exposed to an atmosphere of a different gas. In such a-case difi'erential pressure, or diffusion of gases at same pressure inside of the.holder and outside the same are depended upon to change the gas content and hence to 40 detailed description, the invention is not, of

change the ionic current flow.

While certain specific embodiments of the invention have been selected for illustration and course, limited in its application to these embodiments. The embodiments described should be taken as illustrative of the invention and not as restrictive thereof.' 1

What is claimed is:

1. In a device fortesting the effectiveness of the seal of a closed container that has been sealed at a certain gas pressure and has two spaced electrically conductive members extending from the exterior of said container to a point therewithin and constituting a plate for supporting a container being tested, said plate being provided sealed, means for applying a regulated voltage across said conductive members while maintaining the pressure at the reduced value, and means for observing the magnitude of resulting ionic current fiow between the respective portions of said conductive members within the container: and means for so spacing each container from the adjacent portion of the face of said plate that n a passage is defined by the adjacent surfaces of must be considered when making the tests, as-

- members being in cooperative position with respect to said mounting plate whereby each pair ofsaid contact members engages the external terminals of a respective one of the containers being tested, and means including recessed shoulders engaging said external terminal for so spacing the containers being tested from said plate that a restricted passage is defined between said terminal members by adjacent faces of the containers and the plate of sufiicient size to permit the free flow of air but having a lesser dimension of the order of a millimeter.

3. In a system for testing sealed containers of the type provided with two spaced projecting electrical terminal pin members by a testing procedure involving the application of a voltage across said electrical terminal pin members, means for reducing the gas pressure surrounding said containers and said electrical terminal pin members substantially below atmospheric pressure, means for supporting and connecting into the testing system the containers being tested comprising a plate, a plurality of apertures in said plate adapted to receive said electrical terminal pin members of the containers mounted on said plate for testing each of said apertures comprising three sections linearly arranged and coaxial the first of said three sections being in the form of a relatively shallow cavity the diameter 'of which is appreciably greater than that of the electrical terminal pin members, the second of said three sections being in-the form of a relatively short cylinder the diameter of which is slightly greater than that of said electrical terminal pin members, and the third of said three sections being in the form of a relatively long cylinder the diameter of which is substantially the same as that of the said first section, an electrical contact member positioned in each of the said third sections adapted to receive terminal pin members, means for hermetically closing that end of each of said. third sections farthest from the respective second section, the depth 01 said first aperture sections being so fixed that engagement of said terminal pins therewith will positively space said containers from the nearest portion of said plate a distance of the order of a millimeter.

4. The combination in a device for testing the eflectiveness of the seals of selected ones of a plurality of crystal units enclosed in sealed containers and having spaced terminals projecting therefrom for electrical connection to elements therewithin, of means for receiving a plurality of said crystal units, including a mounting plate having a plurality of recesses therein arranged to receive said spaced terminals of crystal units to be tested, a connecting member disposed within each of said recesses and arranged to engage one of said spaced terminals therein. a cover for said mounting plate and crystal units, and means for sealing said cover about said mounting plate to form therewith an evacuable chamber, of means for spacing each of said units from said mounting plate to, define therewith a path between said spaced terminals not exceeding in height the length of the mean free path of an electron when 18 nection to one of said spaced terminals within said chamber, means for providing external connections to said means for making electrical connection, while providing a tight seal thereabouts.

5. The combination, in a device for testing the seals of a plurality of crystal units having spaced terminals for electrical connection thereto, of a mounting plate arranged to support said units, a cover arranged to form, with said mounting plate, an air-tight chamber, seating apertures formed through said mounting plate, recessed shoulders about said apertures arranged to cooperate with said crystal units to limit the separation between said plate and said units to a maximum of less than the mean free path of electrons within said chamber, enlarged bores in registry with said apertures, resilient members disposed within said apertures and arranged to engage said spaced terminals, means extending externally of said chamber for providing connection to said resilient members, and cooperating means for maintaining an air-tight enclosure about said means for providing connection to said resilient members.

6. In a device for testing the effectiveness of the seals of a plurality of enclosed crystal units having spaced pin-type terminals projecting therefrom, the combination of a mounting plate for receiving such crystal units to be tested, a cover cooperating with said plate to form an airtight chamber, and permitting relatively free passage of air between the interior of those of said units having defective seals and said chamber, and means for preventing ionization curent fiow between said terminals exteriorly of said crystal units.

7. In a device for testing the eflectiveness of the seals of a plurality of enclosed crystal units having spaced pin-type terminals projecting therefrom, the combination of a mounting plate for receiving such crystal units to be tested, a

I cover cooperating with said plate to form an airtight chamber, means comprising recesses formed in said mounting plate. for permitting entry of said terminals to a depth below the surface of said plate defining, between said plate, the ad jacent surface of said unit, and said pin-type terminals, an air leakage space having its lesser REFERENCES CITED The following references are of record in the file of this patent:

UNITED STATES PATENTS Number Name I Date 857,388 Eno June 18, 1907 1,377,282 Schafer May 10, 1921 1.897389 Baker Jan. 1, 1929 1,789,557 Machlett Jan. 20, 1981 1,957,778 Hopkins May 8, 1984 1,988,869 Kimble Dec. 11, 1984 1,984,482 Imamura et al Dec. 18, 1984 2,058,846 Waterman Oct. 27, 1988 2,088,218 Schilling July 27, 1987. 2,169,005 Grleae Aug. 15, 1 2,840,887 Meeker--1 Feb. 1, 1

Images